← home
Github: datasets/msmarco_document.py

ir_datasets: MSMARCO (document)

Index
  1. msmarco-document
  2. msmarco-document/dev
  3. msmarco-document/eval
  4. msmarco-document/orcas
  5. msmarco-document/train
  6. msmarco-document/trec-dl-2019
  7. msmarco-document/trec-dl-2019/judged
  8. msmarco-document/trec-dl-2020
  9. msmarco-document/trec-dl-2020/judged
  10. msmarco-document/trec-dl-hard
  11. msmarco-document/trec-dl-hard/fold1
  12. msmarco-document/trec-dl-hard/fold2
  13. msmarco-document/trec-dl-hard/fold3
  14. msmarco-document/trec-dl-hard/fold4
  15. msmarco-document/trec-dl-hard/fold5

"msmarco-document"

"Based the questions in the [MS-MARCO] Question Answering Dataset and the documents which answered the questions a document ranking task was formulated. There are 3.2 million documents and the goal is to rank based on their relevance. Relevance labels are derived from what passages was marked as having the answer in the QnA dataset."

docs

Language: en

Document type:
MsMarcoDocument: (namedtuple)
  1. doc_id: str
  2. url: str
  3. title: str
  4. body: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, url, title, body>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document docs
[doc_id]    [url]    [title]    [body]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document')
# Index msmarco-document
indexer = pt.IterDictIndexer('./indices/msmarco-document')
index_ref = indexer.index(dataset.get_corpus_iter(), fields=['url', 'title', 'body'])

You can find more details about PyTerrier indexing here.

Citation

ir_datasets.bib:

\cite{Bajaj2016Msmarco}

Bibtex:

@inproceedings{Bajaj2016Msmarco, title={MS MARCO: A Human Generated MAchine Reading COmprehension Dataset}, author={Payal Bajaj, Daniel Campos, Nick Craswell, Li Deng, Jianfeng Gao, Xiaodong Liu, Rangan Majumder, Andrew McNamara, Bhaskar Mitra, Tri Nguyen, Mir Rosenberg, Xia Song, Alina Stoica, Saurabh Tiwary, Tong Wang}, booktitle={InCoCo@NIPS}, year={2016} }

"msmarco-document/dev"

Official dev set. All queries have exactly 1 (positive) relevance judgment.

scoreddocs are the top 100 results from Indri QL. These are used for the "re-ranking" setting.

Official evaluation measures: RR@10

queries

Language: en

Query type:
GenericQuery: (namedtuple)
  1. query_id: str
  2. text: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/dev")
for query in dataset.queries_iter():
    query # namedtuple<query_id, text>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/dev queries
[query_id]    [text]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/dev')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pipeline(dataset.get_topics())

You can find more details about PyTerrier retrieval here.

docs

Language: en

Note: Uses docs from msmarco-document

Document type:
MsMarcoDocument: (namedtuple)
  1. doc_id: str
  2. url: str
  3. title: str
  4. body: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/dev")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, url, title, body>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/dev docs
[doc_id]    [url]    [title]    [body]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/dev')
# Index msmarco-document
indexer = pt.IterDictIndexer('./indices/msmarco-document')
index_ref = indexer.index(dataset.get_corpus_iter(), fields=['url', 'title', 'body'])

You can find more details about PyTerrier indexing here.

qrels
Query relevance judgment type:
TrecQrel: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. relevance: int
  4. iteration: str

Relevance levels

Rel.Definition
1Labeled by crowd worker as relevant

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/dev")
for qrel in dataset.qrels_iter():
    qrel # namedtuple<query_id, doc_id, relevance, iteration>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/dev qrels --format tsv
[query_id]    [doc_id]    [relevance]    [iteration]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
from pyterrier.measures import *
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/dev')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pt.Experiment(
    [pipeline],
    dataset.get_topics(),
    dataset.get_qrels(),
    [RR@10]
)

You can find more details about PyTerrier experiments here.

scoreddocs
Scored Document type:
GenericScoredDoc: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. score: float

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/dev")
for scoreddoc in dataset.scoreddocs_iter():
    scoreddoc # namedtuple<query_id, doc_id, score>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/dev scoreddocs --format tsv
[query_id]    [doc_id]    [score]
...

You can find more details about the CLI here.

PyTerrier

No example available for PyTerrier

Citation

ir_datasets.bib:

\cite{Bajaj2016Msmarco}

Bibtex:

@inproceedings{Bajaj2016Msmarco, title={MS MARCO: A Human Generated MAchine Reading COmprehension Dataset}, author={Payal Bajaj, Daniel Campos, Nick Craswell, Li Deng, Jianfeng Gao, Xiaodong Liu, Rangan Majumder, Andrew McNamara, Bhaskar Mitra, Tri Nguyen, Mir Rosenberg, Xia Song, Alina Stoica, Saurabh Tiwary, Tong Wang}, booktitle={InCoCo@NIPS}, year={2016} }

"msmarco-document/eval"

Official eval set for submission to MS MARCO leaderboard. Relevance judgments are hidden.

scoreddocs are the top 100 results from Indri QL. These are used for the "re-ranking" setting.

Official evaluation measures: RR@10

queries

Language: en

Query type:
GenericQuery: (namedtuple)
  1. query_id: str
  2. text: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/eval")
for query in dataset.queries_iter():
    query # namedtuple<query_id, text>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/eval queries
[query_id]    [text]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/eval')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pipeline(dataset.get_topics())

You can find more details about PyTerrier retrieval here.

docs

Language: en

Note: Uses docs from msmarco-document

Document type:
MsMarcoDocument: (namedtuple)
  1. doc_id: str
  2. url: str
  3. title: str
  4. body: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/eval")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, url, title, body>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/eval docs
[doc_id]    [url]    [title]    [body]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/eval')
# Index msmarco-document
indexer = pt.IterDictIndexer('./indices/msmarco-document')
index_ref = indexer.index(dataset.get_corpus_iter(), fields=['url', 'title', 'body'])

You can find more details about PyTerrier indexing here.

scoreddocs
Scored Document type:
GenericScoredDoc: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. score: float

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/eval")
for scoreddoc in dataset.scoreddocs_iter():
    scoreddoc # namedtuple<query_id, doc_id, score>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/eval scoreddocs --format tsv
[query_id]    [doc_id]    [score]
...

You can find more details about the CLI here.

PyTerrier

No example available for PyTerrier

Citation

ir_datasets.bib:

\cite{Bajaj2016Msmarco}

Bibtex:

@inproceedings{Bajaj2016Msmarco, title={MS MARCO: A Human Generated MAchine Reading COmprehension Dataset}, author={Payal Bajaj, Daniel Campos, Nick Craswell, Li Deng, Jianfeng Gao, Xiaodong Liu, Rangan Majumder, Andrew McNamara, Bhaskar Mitra, Tri Nguyen, Mir Rosenberg, Xia Song, Alina Stoica, Saurabh Tiwary, Tong Wang}, booktitle={InCoCo@NIPS}, year={2016} }

"msmarco-document/orcas"

"ORCAS is a click-based dataset associated with the TREC Deep Learning Track. It covers 1.4 million of the TREC DL documents, providing 18 million connections to 10 million distinct queries."

  • Queries: From query log
  • Relevance Data: User clicks
  • Scored docs: Indri Query Likelihood model
  • Dataset Paper

Official evaluation measures: RR, nDCG

queries

Language: en

Query type:
GenericQuery: (namedtuple)
  1. query_id: str
  2. text: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/orcas")
for query in dataset.queries_iter():
    query # namedtuple<query_id, text>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/orcas queries
[query_id]    [text]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/orcas')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pipeline(dataset.get_topics())

You can find more details about PyTerrier retrieval here.

docs

Language: en

Note: Uses docs from msmarco-document

Document type:
MsMarcoDocument: (namedtuple)
  1. doc_id: str
  2. url: str
  3. title: str
  4. body: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/orcas")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, url, title, body>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/orcas docs
[doc_id]    [url]    [title]    [body]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/orcas')
# Index msmarco-document
indexer = pt.IterDictIndexer('./indices/msmarco-document')
index_ref = indexer.index(dataset.get_corpus_iter(), fields=['url', 'title', 'body'])

You can find more details about PyTerrier indexing here.

qrels
Query relevance judgment type:
TrecQrel: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. relevance: int
  4. iteration: str

Relevance levels

Rel.Definition
1User click

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/orcas")
for qrel in dataset.qrels_iter():
    qrel # namedtuple<query_id, doc_id, relevance, iteration>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/orcas qrels --format tsv
[query_id]    [doc_id]    [relevance]    [iteration]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
from pyterrier.measures import *
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/orcas')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pt.Experiment(
    [pipeline],
    dataset.get_topics(),
    dataset.get_qrels(),
    [RR, nDCG]
)

You can find more details about PyTerrier experiments here.

scoreddocs
Scored Document type:
GenericScoredDoc: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. score: float

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/orcas")
for scoreddoc in dataset.scoreddocs_iter():
    scoreddoc # namedtuple<query_id, doc_id, score>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/orcas scoreddocs --format tsv
[query_id]    [doc_id]    [score]
...

You can find more details about the CLI here.

PyTerrier

No example available for PyTerrier

Citation

ir_datasets.bib:

\cite{Craswell2020Orcas}

Bibtex:

@article{Craswell2020Orcas, title={ORCAS: 18 Million Clicked Query-Document Pairs for Analyzing Search}, author={Craswell, Nick and Campos, Daniel and Mitra, Bhaskar and Yilmaz, Emine and Billerbeck, Bodo}, journal={arXiv preprint arXiv:2006.05324}, year={2020} }

"msmarco-document/train"

Official train set. All queries have exactly 1 (positive) relevance judgment.

scoreddocs are the top 100 results from Indri QL. These are used for the "re-ranking" setting.

Official evaluation measures: RR@10

queries

Language: en

Query type:
GenericQuery: (namedtuple)
  1. query_id: str
  2. text: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/train")
for query in dataset.queries_iter():
    query # namedtuple<query_id, text>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/train queries
[query_id]    [text]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/train')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pipeline(dataset.get_topics())

You can find more details about PyTerrier retrieval here.

docs

Language: en

Note: Uses docs from msmarco-document

Document type:
MsMarcoDocument: (namedtuple)
  1. doc_id: str
  2. url: str
  3. title: str
  4. body: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/train")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, url, title, body>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/train docs
[doc_id]    [url]    [title]    [body]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/train')
# Index msmarco-document
indexer = pt.IterDictIndexer('./indices/msmarco-document')
index_ref = indexer.index(dataset.get_corpus_iter(), fields=['url', 'title', 'body'])

You can find more details about PyTerrier indexing here.

qrels
Query relevance judgment type:
TrecQrel: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. relevance: int
  4. iteration: str

Relevance levels

Rel.Definition
1Labeled by crowd worker as relevant

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/train")
for qrel in dataset.qrels_iter():
    qrel # namedtuple<query_id, doc_id, relevance, iteration>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/train qrels --format tsv
[query_id]    [doc_id]    [relevance]    [iteration]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
from pyterrier.measures import *
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/train')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pt.Experiment(
    [pipeline],
    dataset.get_topics(),
    dataset.get_qrels(),
    [RR@10]
)

You can find more details about PyTerrier experiments here.

scoreddocs
Scored Document type:
GenericScoredDoc: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. score: float

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/train")
for scoreddoc in dataset.scoreddocs_iter():
    scoreddoc # namedtuple<query_id, doc_id, score>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/train scoreddocs --format tsv
[query_id]    [doc_id]    [score]
...

You can find more details about the CLI here.

PyTerrier

No example available for PyTerrier

Citation

ir_datasets.bib:

\cite{Bajaj2016Msmarco}

Bibtex:

@inproceedings{Bajaj2016Msmarco, title={MS MARCO: A Human Generated MAchine Reading COmprehension Dataset}, author={Payal Bajaj, Daniel Campos, Nick Craswell, Li Deng, Jianfeng Gao, Xiaodong Liu, Rangan Majumder, Andrew McNamara, Bhaskar Mitra, Tri Nguyen, Mir Rosenberg, Xia Song, Alina Stoica, Saurabh Tiwary, Tong Wang}, booktitle={InCoCo@NIPS}, year={2016} }

"msmarco-document/trec-dl-2019"

Queries from the TREC Deep Learning (DL) 2019 shared task, which were sampled from msmarco-document/eval. A subset of these queries were judged by NIST assessors, (filtered list available in msmarco-document/trec-dl-2019/judged).

Official evaluation measures: nDCG@10, RR, MAP

queries

Language: en

Query type:
GenericQuery: (namedtuple)
  1. query_id: str
  2. text: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-2019")
for query in dataset.queries_iter():
    query # namedtuple<query_id, text>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-2019 queries
[query_id]    [text]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-2019')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pipeline(dataset.get_topics())

You can find more details about PyTerrier retrieval here.

docs

Language: en

Note: Uses docs from msmarco-document

Document type:
MsMarcoDocument: (namedtuple)
  1. doc_id: str
  2. url: str
  3. title: str
  4. body: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-2019")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, url, title, body>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-2019 docs
[doc_id]    [url]    [title]    [body]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-2019')
# Index msmarco-document
indexer = pt.IterDictIndexer('./indices/msmarco-document')
index_ref = indexer.index(dataset.get_corpus_iter(), fields=['url', 'title', 'body'])

You can find more details about PyTerrier indexing here.

qrels
Query relevance judgment type:
TrecQrel: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. relevance: int
  4. iteration: str

Relevance levels

Rel.Definition
0Irrelevant: Document does not provide any useful information about the query
1Relevant: Document provides some information relevant to the query, which may be minimal.
2Highly relevant: The content of this document provides substantial information on the query.
3Perfectly relevant: Document is dedicated to the query, it is worthy of being a top result in a search engine.

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-2019")
for qrel in dataset.qrels_iter():
    qrel # namedtuple<query_id, doc_id, relevance, iteration>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-2019 qrels --format tsv
[query_id]    [doc_id]    [relevance]    [iteration]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
from pyterrier.measures import *
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-2019')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pt.Experiment(
    [pipeline],
    dataset.get_topics(),
    dataset.get_qrels(),
    [nDCG@10, RR, MAP]
)

You can find more details about PyTerrier experiments here.

scoreddocs
Scored Document type:
GenericScoredDoc: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. score: float

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-2019")
for scoreddoc in dataset.scoreddocs_iter():
    scoreddoc # namedtuple<query_id, doc_id, score>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-2019 scoreddocs --format tsv
[query_id]    [doc_id]    [score]
...

You can find more details about the CLI here.

PyTerrier

No example available for PyTerrier

Citation

ir_datasets.bib:

\cite{Craswell2019TrecDl,Bajaj2016Msmarco}

Bibtex:

@inproceedings{Craswell2019TrecDl, title={Overview of the TREC 2019 deep learning track}, author={Nick Craswell and Bhaskar Mitra and Emine Yilmaz and Daniel Campos and Ellen Voorhees}, booktitle={TREC 2019}, year={2019} } @inproceedings{Bajaj2016Msmarco, title={MS MARCO: A Human Generated MAchine Reading COmprehension Dataset}, author={Payal Bajaj, Daniel Campos, Nick Craswell, Li Deng, Jianfeng Gao, Xiaodong Liu, Rangan Majumder, Andrew McNamara, Bhaskar Mitra, Tri Nguyen, Mir Rosenberg, Xia Song, Alina Stoica, Saurabh Tiwary, Tong Wang}, booktitle={InCoCo@NIPS}, year={2016} }

"msmarco-document/trec-dl-2019/judged"

Subset of msmarco-document/trec-dl-2019, only including queries with qrels.

Official evaluation measures: nDCG@10, RR, MAP

queries

Language: en

Query type:
GenericQuery: (namedtuple)
  1. query_id: str
  2. text: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-2019/judged")
for query in dataset.queries_iter():
    query # namedtuple<query_id, text>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-2019/judged queries
[query_id]    [text]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-2019/judged')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pipeline(dataset.get_topics())

You can find more details about PyTerrier retrieval here.

docs

Language: en

Note: Uses docs from msmarco-document

Document type:
MsMarcoDocument: (namedtuple)
  1. doc_id: str
  2. url: str
  3. title: str
  4. body: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-2019/judged")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, url, title, body>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-2019/judged docs
[doc_id]    [url]    [title]    [body]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-2019/judged')
# Index msmarco-document
indexer = pt.IterDictIndexer('./indices/msmarco-document')
index_ref = indexer.index(dataset.get_corpus_iter(), fields=['url', 'title', 'body'])

You can find more details about PyTerrier indexing here.

qrels
Query relevance judgment type:
TrecQrel: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. relevance: int
  4. iteration: str

Relevance levels

Rel.Definition
0Irrelevant: Document does not provide any useful information about the query
1Relevant: Document provides some information relevant to the query, which may be minimal.
2Highly relevant: The content of this document provides substantial information on the query.
3Perfectly relevant: Document is dedicated to the query, it is worthy of being a top result in a search engine.

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-2019/judged")
for qrel in dataset.qrels_iter():
    qrel # namedtuple<query_id, doc_id, relevance, iteration>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-2019/judged qrels --format tsv
[query_id]    [doc_id]    [relevance]    [iteration]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
from pyterrier.measures import *
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-2019/judged')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pt.Experiment(
    [pipeline],
    dataset.get_topics(),
    dataset.get_qrels(),
    [nDCG@10, RR, MAP]
)

You can find more details about PyTerrier experiments here.

scoreddocs
Scored Document type:
GenericScoredDoc: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. score: float

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-2019/judged")
for scoreddoc in dataset.scoreddocs_iter():
    scoreddoc # namedtuple<query_id, doc_id, score>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-2019/judged scoreddocs --format tsv
[query_id]    [doc_id]    [score]
...

You can find more details about the CLI here.

PyTerrier

No example available for PyTerrier

Citation

ir_datasets.bib:

\cite{Craswell2019TrecDl,Bajaj2016Msmarco}

Bibtex:

@inproceedings{Craswell2019TrecDl, title={Overview of the TREC 2019 deep learning track}, author={Nick Craswell and Bhaskar Mitra and Emine Yilmaz and Daniel Campos and Ellen Voorhees}, booktitle={TREC 2019}, year={2019} } @inproceedings{Bajaj2016Msmarco, title={MS MARCO: A Human Generated MAchine Reading COmprehension Dataset}, author={Payal Bajaj, Daniel Campos, Nick Craswell, Li Deng, Jianfeng Gao, Xiaodong Liu, Rangan Majumder, Andrew McNamara, Bhaskar Mitra, Tri Nguyen, Mir Rosenberg, Xia Song, Alina Stoica, Saurabh Tiwary, Tong Wang}, booktitle={InCoCo@NIPS}, year={2016} }

"msmarco-document/trec-dl-2020"

Queries from the TREC Deep Learning (DL) 2020 shared task, which were sampled from msmarco-document/eval. A subset of these queries were judged by NIST assessors, (filtered list available in msmarco-document/trec-dl-2020/judged).

Official evaluation measures: nDCG@10, RR, MAP

queries

Language: en

Query type:
GenericQuery: (namedtuple)
  1. query_id: str
  2. text: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-2020")
for query in dataset.queries_iter():
    query # namedtuple<query_id, text>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-2020 queries
[query_id]    [text]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-2020')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pipeline(dataset.get_topics())

You can find more details about PyTerrier retrieval here.

docs

Language: en

Note: Uses docs from msmarco-document

Document type:
MsMarcoDocument: (namedtuple)
  1. doc_id: str
  2. url: str
  3. title: str
  4. body: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-2020")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, url, title, body>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-2020 docs
[doc_id]    [url]    [title]    [body]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-2020')
# Index msmarco-document
indexer = pt.IterDictIndexer('./indices/msmarco-document')
index_ref = indexer.index(dataset.get_corpus_iter(), fields=['url', 'title', 'body'])

You can find more details about PyTerrier indexing here.

qrels
Query relevance judgment type:
TrecQrel: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. relevance: int
  4. iteration: str

Relevance levels

Rel.Definition
0Irrelevant: Document does not provide any useful information about the query
1Relevant: Document provides some information relevant to the query, which may be minimal.
2Highly relevant: The content of this document provides substantial information on the query.
3Perfectly relevant: Document is dedicated to the query, it is worthy of being a top result in a search engine.

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-2020")
for qrel in dataset.qrels_iter():
    qrel # namedtuple<query_id, doc_id, relevance, iteration>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-2020 qrels --format tsv
[query_id]    [doc_id]    [relevance]    [iteration]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
from pyterrier.measures import *
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-2020')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pt.Experiment(
    [pipeline],
    dataset.get_topics(),
    dataset.get_qrels(),
    [nDCG@10, RR, MAP]
)

You can find more details about PyTerrier experiments here.

scoreddocs
Scored Document type:
GenericScoredDoc: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. score: float

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-2020")
for scoreddoc in dataset.scoreddocs_iter():
    scoreddoc # namedtuple<query_id, doc_id, score>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-2020 scoreddocs --format tsv
[query_id]    [doc_id]    [score]
...

You can find more details about the CLI here.

PyTerrier

No example available for PyTerrier

Citation

ir_datasets.bib:

\cite{Craswell2020TrecDl,Bajaj2016Msmarco}

Bibtex:

@inproceedings{Craswell2020TrecDl, title={Overview of the TREC 2020 deep learning track}, author={Nick Craswell and Bhaskar Mitra and Emine Yilmaz and Daniel Campos}, booktitle={TREC}, year={2020} } @inproceedings{Bajaj2016Msmarco, title={MS MARCO: A Human Generated MAchine Reading COmprehension Dataset}, author={Payal Bajaj, Daniel Campos, Nick Craswell, Li Deng, Jianfeng Gao, Xiaodong Liu, Rangan Majumder, Andrew McNamara, Bhaskar Mitra, Tri Nguyen, Mir Rosenberg, Xia Song, Alina Stoica, Saurabh Tiwary, Tong Wang}, booktitle={InCoCo@NIPS}, year={2016} }

"msmarco-document/trec-dl-2020/judged"

Subset of msmarco-document/trec-dl-2020, only including queries with qrels.

Official evaluation measures: nDCG@10, RR, MAP

queries

Language: en

Query type:
GenericQuery: (namedtuple)
  1. query_id: str
  2. text: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-2020/judged")
for query in dataset.queries_iter():
    query # namedtuple<query_id, text>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-2020/judged queries
[query_id]    [text]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-2020/judged')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pipeline(dataset.get_topics())

You can find more details about PyTerrier retrieval here.

docs

Language: en

Note: Uses docs from msmarco-document

Document type:
MsMarcoDocument: (namedtuple)
  1. doc_id: str
  2. url: str
  3. title: str
  4. body: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-2020/judged")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, url, title, body>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-2020/judged docs
[doc_id]    [url]    [title]    [body]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-2020/judged')
# Index msmarco-document
indexer = pt.IterDictIndexer('./indices/msmarco-document')
index_ref = indexer.index(dataset.get_corpus_iter(), fields=['url', 'title', 'body'])

You can find more details about PyTerrier indexing here.

qrels
Query relevance judgment type:
TrecQrel: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. relevance: int
  4. iteration: str

Relevance levels

Rel.Definition
0Irrelevant: Document does not provide any useful information about the query
1Relevant: Document provides some information relevant to the query, which may be minimal.
2Highly relevant: The content of this document provides substantial information on the query.
3Perfectly relevant: Document is dedicated to the query, it is worthy of being a top result in a search engine.

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-2020/judged")
for qrel in dataset.qrels_iter():
    qrel # namedtuple<query_id, doc_id, relevance, iteration>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-2020/judged qrels --format tsv
[query_id]    [doc_id]    [relevance]    [iteration]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
from pyterrier.measures import *
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-2020/judged')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pt.Experiment(
    [pipeline],
    dataset.get_topics(),
    dataset.get_qrels(),
    [nDCG@10, RR, MAP]
)

You can find more details about PyTerrier experiments here.

scoreddocs
Scored Document type:
GenericScoredDoc: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. score: float

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-2020/judged")
for scoreddoc in dataset.scoreddocs_iter():
    scoreddoc # namedtuple<query_id, doc_id, score>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-2020/judged scoreddocs --format tsv
[query_id]    [doc_id]    [score]
...

You can find more details about the CLI here.

PyTerrier

No example available for PyTerrier

Citation

ir_datasets.bib:

\cite{Craswell2020TrecDl,Bajaj2016Msmarco}

Bibtex:

@inproceedings{Craswell2020TrecDl, title={Overview of the TREC 2020 deep learning track}, author={Nick Craswell and Bhaskar Mitra and Emine Yilmaz and Daniel Campos}, booktitle={TREC}, year={2020} } @inproceedings{Bajaj2016Msmarco, title={MS MARCO: A Human Generated MAchine Reading COmprehension Dataset}, author={Payal Bajaj, Daniel Campos, Nick Craswell, Li Deng, Jianfeng Gao, Xiaodong Liu, Rangan Majumder, Andrew McNamara, Bhaskar Mitra, Tri Nguyen, Mir Rosenberg, Xia Song, Alina Stoica, Saurabh Tiwary, Tong Wang}, booktitle={InCoCo@NIPS}, year={2016} }

"msmarco-document/trec-dl-hard"

A more challenging subset of msmarco-document/trec-dl-2019 and msmarco-document/trec-dl-2020.

Official evaluation measures: nDCG@10, RR(rel=2)

queries

Language: en

Query type:
GenericQuery: (namedtuple)
  1. query_id: str
  2. text: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-hard")
for query in dataset.queries_iter():
    query # namedtuple<query_id, text>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-hard queries
[query_id]    [text]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-hard')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pipeline(dataset.get_topics())

You can find more details about PyTerrier retrieval here.

docs

Language: en

Note: Uses docs from msmarco-document

Document type:
MsMarcoDocument: (namedtuple)
  1. doc_id: str
  2. url: str
  3. title: str
  4. body: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-hard")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, url, title, body>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-hard docs
[doc_id]    [url]    [title]    [body]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-hard')
# Index msmarco-document
indexer = pt.IterDictIndexer('./indices/msmarco-document')
index_ref = indexer.index(dataset.get_corpus_iter(), fields=['url', 'title', 'body'])

You can find more details about PyTerrier indexing here.

qrels
Query relevance judgment type:
TrecQrel: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. relevance: int
  4. iteration: str

Relevance levels

Rel.Definition
0Irrelevant: Document does not provide any useful information about the query
1Relevant: Document provides some information relevant to the query, which may be minimal.
2Highly relevant: The content of this document provides substantial information on the query.
3Perfectly relevant: Document is dedicated to the query, it is worthy of being a top result in a search engine.

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-hard")
for qrel in dataset.qrels_iter():
    qrel # namedtuple<query_id, doc_id, relevance, iteration>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-hard qrels --format tsv
[query_id]    [doc_id]    [relevance]    [iteration]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
from pyterrier.measures import *
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-hard')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pt.Experiment(
    [pipeline],
    dataset.get_topics(),
    dataset.get_qrels(),
    [nDCG@10, RR(rel=2)]
)

You can find more details about PyTerrier experiments here.

Citation

ir_datasets.bib:

\cite{Mackie2021DlHard,Bajaj2016Msmarco}

Bibtex:

@article{Mackie2021DlHard, title={How Deep is your Learning: the DL-HARD Annotated Deep Learning Dataset}, author={Iain Mackie and Jeffrey Dalton and Andrew Yates}, journal={ArXiv}, year={2021}, volume={abs/2105.07975} } @inproceedings{Bajaj2016Msmarco, title={MS MARCO: A Human Generated MAchine Reading COmprehension Dataset}, author={Payal Bajaj, Daniel Campos, Nick Craswell, Li Deng, Jianfeng Gao, Xiaodong Liu, Rangan Majumder, Andrew McNamara, Bhaskar Mitra, Tri Nguyen, Mir Rosenberg, Xia Song, Alina Stoica, Saurabh Tiwary, Tong Wang}, booktitle={InCoCo@NIPS}, year={2016} }

"msmarco-document/trec-dl-hard/fold1"

A more challenging subset of msmarco-document/trec-dl-2019 and msmarco-document/trec-dl-2020.

Official evaluation measures: nDCG@10, RR(rel=2)

queries

Language: en

Query type:
GenericQuery: (namedtuple)
  1. query_id: str
  2. text: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-hard/fold1")
for query in dataset.queries_iter():
    query # namedtuple<query_id, text>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-hard/fold1 queries
[query_id]    [text]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-hard/fold1')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pipeline(dataset.get_topics())

You can find more details about PyTerrier retrieval here.

docs

Language: en

Note: Uses docs from msmarco-document

Document type:
MsMarcoDocument: (namedtuple)
  1. doc_id: str
  2. url: str
  3. title: str
  4. body: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-hard/fold1")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, url, title, body>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-hard/fold1 docs
[doc_id]    [url]    [title]    [body]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-hard/fold1')
# Index msmarco-document
indexer = pt.IterDictIndexer('./indices/msmarco-document')
index_ref = indexer.index(dataset.get_corpus_iter(), fields=['url', 'title', 'body'])

You can find more details about PyTerrier indexing here.

qrels
Query relevance judgment type:
TrecQrel: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. relevance: int
  4. iteration: str

Relevance levels

Rel.Definition
0Irrelevant: Document does not provide any useful information about the query
1Relevant: Document provides some information relevant to the query, which may be minimal.
2Highly relevant: The content of this document provides substantial information on the query.
3Perfectly relevant: Document is dedicated to the query, it is worthy of being a top result in a search engine.

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-hard/fold1")
for qrel in dataset.qrels_iter():
    qrel # namedtuple<query_id, doc_id, relevance, iteration>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-hard/fold1 qrels --format tsv
[query_id]    [doc_id]    [relevance]    [iteration]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
from pyterrier.measures import *
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-hard/fold1')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pt.Experiment(
    [pipeline],
    dataset.get_topics(),
    dataset.get_qrels(),
    [nDCG@10, RR(rel=2)]
)

You can find more details about PyTerrier experiments here.

Citation

ir_datasets.bib:

\cite{Mackie2021DlHard,Bajaj2016Msmarco}

Bibtex:

@article{Mackie2021DlHard, title={How Deep is your Learning: the DL-HARD Annotated Deep Learning Dataset}, author={Iain Mackie and Jeffrey Dalton and Andrew Yates}, journal={ArXiv}, year={2021}, volume={abs/2105.07975} } @inproceedings{Bajaj2016Msmarco, title={MS MARCO: A Human Generated MAchine Reading COmprehension Dataset}, author={Payal Bajaj, Daniel Campos, Nick Craswell, Li Deng, Jianfeng Gao, Xiaodong Liu, Rangan Majumder, Andrew McNamara, Bhaskar Mitra, Tri Nguyen, Mir Rosenberg, Xia Song, Alina Stoica, Saurabh Tiwary, Tong Wang}, booktitle={InCoCo@NIPS}, year={2016} }

"msmarco-document/trec-dl-hard/fold2"

A more challenging subset of msmarco-document/trec-dl-2019 and msmarco-document/trec-dl-2020.

Official evaluation measures: nDCG@10, RR(rel=2)

queries

Language: en

Query type:
GenericQuery: (namedtuple)
  1. query_id: str
  2. text: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-hard/fold2")
for query in dataset.queries_iter():
    query # namedtuple<query_id, text>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-hard/fold2 queries
[query_id]    [text]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-hard/fold2')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pipeline(dataset.get_topics())

You can find more details about PyTerrier retrieval here.

docs

Language: en

Note: Uses docs from msmarco-document

Document type:
MsMarcoDocument: (namedtuple)
  1. doc_id: str
  2. url: str
  3. title: str
  4. body: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-hard/fold2")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, url, title, body>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-hard/fold2 docs
[doc_id]    [url]    [title]    [body]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-hard/fold2')
# Index msmarco-document
indexer = pt.IterDictIndexer('./indices/msmarco-document')
index_ref = indexer.index(dataset.get_corpus_iter(), fields=['url', 'title', 'body'])

You can find more details about PyTerrier indexing here.

qrels
Query relevance judgment type:
TrecQrel: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. relevance: int
  4. iteration: str

Relevance levels

Rel.Definition
0Irrelevant: Document does not provide any useful information about the query
1Relevant: Document provides some information relevant to the query, which may be minimal.
2Highly relevant: The content of this document provides substantial information on the query.
3Perfectly relevant: Document is dedicated to the query, it is worthy of being a top result in a search engine.

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-hard/fold2")
for qrel in dataset.qrels_iter():
    qrel # namedtuple<query_id, doc_id, relevance, iteration>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-hard/fold2 qrels --format tsv
[query_id]    [doc_id]    [relevance]    [iteration]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
from pyterrier.measures import *
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-hard/fold2')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pt.Experiment(
    [pipeline],
    dataset.get_topics(),
    dataset.get_qrels(),
    [nDCG@10, RR(rel=2)]
)

You can find more details about PyTerrier experiments here.

Citation

ir_datasets.bib:

\cite{Mackie2021DlHard,Bajaj2016Msmarco}

Bibtex:

@article{Mackie2021DlHard, title={How Deep is your Learning: the DL-HARD Annotated Deep Learning Dataset}, author={Iain Mackie and Jeffrey Dalton and Andrew Yates}, journal={ArXiv}, year={2021}, volume={abs/2105.07975} } @inproceedings{Bajaj2016Msmarco, title={MS MARCO: A Human Generated MAchine Reading COmprehension Dataset}, author={Payal Bajaj, Daniel Campos, Nick Craswell, Li Deng, Jianfeng Gao, Xiaodong Liu, Rangan Majumder, Andrew McNamara, Bhaskar Mitra, Tri Nguyen, Mir Rosenberg, Xia Song, Alina Stoica, Saurabh Tiwary, Tong Wang}, booktitle={InCoCo@NIPS}, year={2016} }

"msmarco-document/trec-dl-hard/fold3"

A more challenging subset of msmarco-document/trec-dl-2019 and msmarco-document/trec-dl-2020.

Official evaluation measures: nDCG@10, RR(rel=2)

queries

Language: en

Query type:
GenericQuery: (namedtuple)
  1. query_id: str
  2. text: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-hard/fold3")
for query in dataset.queries_iter():
    query # namedtuple<query_id, text>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-hard/fold3 queries
[query_id]    [text]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-hard/fold3')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pipeline(dataset.get_topics())

You can find more details about PyTerrier retrieval here.

docs

Language: en

Note: Uses docs from msmarco-document

Document type:
MsMarcoDocument: (namedtuple)
  1. doc_id: str
  2. url: str
  3. title: str
  4. body: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-hard/fold3")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, url, title, body>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-hard/fold3 docs
[doc_id]    [url]    [title]    [body]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-hard/fold3')
# Index msmarco-document
indexer = pt.IterDictIndexer('./indices/msmarco-document')
index_ref = indexer.index(dataset.get_corpus_iter(), fields=['url', 'title', 'body'])

You can find more details about PyTerrier indexing here.

qrels
Query relevance judgment type:
TrecQrel: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. relevance: int
  4. iteration: str

Relevance levels

Rel.Definition
0Irrelevant: Document does not provide any useful information about the query
1Relevant: Document provides some information relevant to the query, which may be minimal.
2Highly relevant: The content of this document provides substantial information on the query.
3Perfectly relevant: Document is dedicated to the query, it is worthy of being a top result in a search engine.

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-hard/fold3")
for qrel in dataset.qrels_iter():
    qrel # namedtuple<query_id, doc_id, relevance, iteration>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-hard/fold3 qrels --format tsv
[query_id]    [doc_id]    [relevance]    [iteration]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
from pyterrier.measures import *
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-hard/fold3')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pt.Experiment(
    [pipeline],
    dataset.get_topics(),
    dataset.get_qrels(),
    [nDCG@10, RR(rel=2)]
)

You can find more details about PyTerrier experiments here.

Citation

ir_datasets.bib:

\cite{Mackie2021DlHard,Bajaj2016Msmarco}

Bibtex:

@article{Mackie2021DlHard, title={How Deep is your Learning: the DL-HARD Annotated Deep Learning Dataset}, author={Iain Mackie and Jeffrey Dalton and Andrew Yates}, journal={ArXiv}, year={2021}, volume={abs/2105.07975} } @inproceedings{Bajaj2016Msmarco, title={MS MARCO: A Human Generated MAchine Reading COmprehension Dataset}, author={Payal Bajaj, Daniel Campos, Nick Craswell, Li Deng, Jianfeng Gao, Xiaodong Liu, Rangan Majumder, Andrew McNamara, Bhaskar Mitra, Tri Nguyen, Mir Rosenberg, Xia Song, Alina Stoica, Saurabh Tiwary, Tong Wang}, booktitle={InCoCo@NIPS}, year={2016} }

"msmarco-document/trec-dl-hard/fold4"

A more challenging subset of msmarco-document/trec-dl-2019 and msmarco-document/trec-dl-2020.

Official evaluation measures: nDCG@10, RR(rel=2)

queries

Language: en

Query type:
GenericQuery: (namedtuple)
  1. query_id: str
  2. text: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-hard/fold4")
for query in dataset.queries_iter():
    query # namedtuple<query_id, text>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-hard/fold4 queries
[query_id]    [text]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-hard/fold4')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pipeline(dataset.get_topics())

You can find more details about PyTerrier retrieval here.

docs

Language: en

Note: Uses docs from msmarco-document

Document type:
MsMarcoDocument: (namedtuple)
  1. doc_id: str
  2. url: str
  3. title: str
  4. body: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-hard/fold4")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, url, title, body>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-hard/fold4 docs
[doc_id]    [url]    [title]    [body]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-hard/fold4')
# Index msmarco-document
indexer = pt.IterDictIndexer('./indices/msmarco-document')
index_ref = indexer.index(dataset.get_corpus_iter(), fields=['url', 'title', 'body'])

You can find more details about PyTerrier indexing here.

qrels
Query relevance judgment type:
TrecQrel: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. relevance: int
  4. iteration: str

Relevance levels

Rel.Definition
0Irrelevant: Document does not provide any useful information about the query
1Relevant: Document provides some information relevant to the query, which may be minimal.
2Highly relevant: The content of this document provides substantial information on the query.
3Perfectly relevant: Document is dedicated to the query, it is worthy of being a top result in a search engine.

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-hard/fold4")
for qrel in dataset.qrels_iter():
    qrel # namedtuple<query_id, doc_id, relevance, iteration>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-hard/fold4 qrels --format tsv
[query_id]    [doc_id]    [relevance]    [iteration]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
from pyterrier.measures import *
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-hard/fold4')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pt.Experiment(
    [pipeline],
    dataset.get_topics(),
    dataset.get_qrels(),
    [nDCG@10, RR(rel=2)]
)

You can find more details about PyTerrier experiments here.

Citation

ir_datasets.bib:

\cite{Mackie2021DlHard,Bajaj2016Msmarco}

Bibtex:

@article{Mackie2021DlHard, title={How Deep is your Learning: the DL-HARD Annotated Deep Learning Dataset}, author={Iain Mackie and Jeffrey Dalton and Andrew Yates}, journal={ArXiv}, year={2021}, volume={abs/2105.07975} } @inproceedings{Bajaj2016Msmarco, title={MS MARCO: A Human Generated MAchine Reading COmprehension Dataset}, author={Payal Bajaj, Daniel Campos, Nick Craswell, Li Deng, Jianfeng Gao, Xiaodong Liu, Rangan Majumder, Andrew McNamara, Bhaskar Mitra, Tri Nguyen, Mir Rosenberg, Xia Song, Alina Stoica, Saurabh Tiwary, Tong Wang}, booktitle={InCoCo@NIPS}, year={2016} }

"msmarco-document/trec-dl-hard/fold5"

A more challenging subset of msmarco-document/trec-dl-2019 and msmarco-document/trec-dl-2020.

Official evaluation measures: nDCG@10, RR(rel=2)

queries

Language: en

Query type:
GenericQuery: (namedtuple)
  1. query_id: str
  2. text: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-hard/fold5")
for query in dataset.queries_iter():
    query # namedtuple<query_id, text>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-hard/fold5 queries
[query_id]    [text]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-hard/fold5')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pipeline(dataset.get_topics())

You can find more details about PyTerrier retrieval here.

docs

Language: en

Note: Uses docs from msmarco-document

Document type:
MsMarcoDocument: (namedtuple)
  1. doc_id: str
  2. url: str
  3. title: str
  4. body: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-hard/fold5")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, url, title, body>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-hard/fold5 docs
[doc_id]    [url]    [title]    [body]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-hard/fold5')
# Index msmarco-document
indexer = pt.IterDictIndexer('./indices/msmarco-document')
index_ref = indexer.index(dataset.get_corpus_iter(), fields=['url', 'title', 'body'])

You can find more details about PyTerrier indexing here.

qrels
Query relevance judgment type:
TrecQrel: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. relevance: int
  4. iteration: str

Relevance levels

Rel.Definition
0Irrelevant: Document does not provide any useful information about the query
1Relevant: Document provides some information relevant to the query, which may be minimal.
2Highly relevant: The content of this document provides substantial information on the query.
3Perfectly relevant: Document is dedicated to the query, it is worthy of being a top result in a search engine.

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("msmarco-document/trec-dl-hard/fold5")
for qrel in dataset.qrels_iter():
    qrel # namedtuple<query_id, doc_id, relevance, iteration>

You can find more details about the Python API here.

CLI 
ir_datasets export msmarco-document/trec-dl-hard/fold5 qrels --format tsv
[query_id]    [doc_id]    [relevance]    [iteration]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
from pyterrier.measures import *
pt.init()
dataset = pt.get_dataset('irds:msmarco-document/trec-dl-hard/fold5')
index_ref = pt.IndexRef.of('./indices/msmarco-document') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pt.Experiment(
    [pipeline],
    dataset.get_topics(),
    dataset.get_qrels(),
    [nDCG@10, RR(rel=2)]
)

You can find more details about PyTerrier experiments here.

Citation

ir_datasets.bib:

\cite{Mackie2021DlHard,Bajaj2016Msmarco}

Bibtex:

@article{Mackie2021DlHard, title={How Deep is your Learning: the DL-HARD Annotated Deep Learning Dataset}, author={Iain Mackie and Jeffrey Dalton and Andrew Yates}, journal={ArXiv}, year={2021}, volume={abs/2105.07975} } @inproceedings{Bajaj2016Msmarco, title={MS MARCO: A Human Generated MAchine Reading COmprehension Dataset}, author={Payal Bajaj, Daniel Campos, Nick Craswell, Li Deng, Jianfeng Gao, Xiaodong Liu, Rangan Majumder, Andrew McNamara, Bhaskar Mitra, Tri Nguyen, Mir Rosenberg, Xia Song, Alina Stoica, Saurabh Tiwary, Tong Wang}, booktitle={InCoCo@NIPS}, year={2016} }